Success Rates in Protein Crystallography
Overall success rates (Thornton):
- Obtain milligrams of pure, soluble protein.
- About 60% of sequences express well.
- Only one half (prokaryotic) to one quarter (eukaryotic) of these are soluble.
- Obtain high quality crystals.
- Typically hundreds of crystallization
conditions are tested.
- Crystals must be singular, sufficiently large,
and preferably neither needles nor plates.
- Crystals with large numbers of copies in the asymmetric unit
- About half of highly expressed soluble proteins crystallize,
but only about one third of these crystals are suitable.
By April 2007, of ~100,000 targets cloned in Structural Genomics Projects,
4,500 had been solved
Of these, ~2,000 nonredundant solved targets represent
12% of the
~16,000-target goal of Structural Genomics.
- Possibly up to
Eric Martz, University of Massachusetts, July 2003 (revised February 2004, July 2005,
June 2006, April 2007)
Thanks to Byron Rubin for some insights here.
- TargetDB maintained by the
RCSB can be queried by Status to find the current
numbers of targets that have reached each stage. About 25% of completed
results are being solved by NMR.
Structural genomics takes off, Thornton,
Trends Biochem. Sci., 26:88, 2001.
genomics. Tapping DNA for structures produces a trickle,
Service, Science 298:948, 2002.
genomics: current progress, Gerstein et al.,
Science 299:1663, 2003.
The authors make the point that overall
progress is much larger than the number of solved structures, since each
solution serves as a template for homology modeling a large family of
sequences. Also, work is often stopped on a target when a sequence-related
target is solved; hence not all uncompleted targets are "failures".